Introduction

Lipid profile as estimated by Serum Total Cholesterol, Triglycerides, HDL. LDL indicates general status of circulating lipid concentrations. When we want to assess total body fat either in Obesity, Metabolic syndrome or any other endocrine disorder other methods available are - waist circumference (WC), waist hip ratio (WHR), Body mass index (BMI). Recently assessment of Body fat percentage (BFP) is gaining popularity over these anthropometric data. Body fat is basically essential fat and storage fat. Essential fat is necessary to maintain life and reproductive functions. Storage fat accumulates in adipose tissue & has important function of protecting intrathoracic & intraabdominal organs.

The following methods are available to assess BFP

Most of these methods need instruments which may not be available and moreover accuracy of the values obtained needs to be cross-checked. Anthropometric measurements include neck circumference (NC), hip circumference (HC), WHR (waist hip ratio), WC (waist circumference) & are markers for central obesity assessment. BMI cannot distinguish between body fat mass and lean body mass and BMI varies with Ht, weight and age. BMI raises if skeletal muscle mass is more or when bones are larger. To overcome the hurdle, BFP was being considered. Computed Tomography (CT) or dual-energy X-ray absorptiometry (DXA) are frequently used methods for evaluating the distribution of body fat.1

Most obesity indices such as NC, HC, WHR, WC are only markers of central obesity. BF% is a measure of fitness level irrespective of Ht. & Wt. BFP is total body fat divided by total body mass. BFP was calculated by the formula given below.

BFP = 1.20 × BMI + 0.23 × Age - 16.2 (♂) BFP = 1.20 × BMI + 0.23 × Age - 5.4 (♀)

The ideal BFP for men and women were as under.

Type-2 Diabetes Mellitus (T2DM) accounts for 23 deaths per lakh population, which translates into more than 3 lakh deaths per year (or 3% of all deaths) in India.2 T2DM frequently leads to Metabolic syndrome(MetS) and complications thereof.

The aim of this study was to find correlation of body fat percentage (BF%) and Body mass index (BMI) with lipid profile in Diabetes Mellitus patients for the prediction of Metabolic Syndrome.

Aim and Objectives

To find correlation of BMI and BFP with lipid profile in type 2 DM and assess advantage of BFP over BMI in evaluating risk of MetS. The objective of the study was to find the relation between BMI and BFP.

Materials and Methods

The cross- sectional study was conducted in Government General Hospital, Machilipatnam in Dec/Jan 2022/23. Based on inclusion and exclusion criteria, 35 Type 2 Diabetes Mellitus (T2DM) patients both males and females in the age group of 30yrs -55yrs and 35 age and sex matched healthy controls were selected for the study after obtaining their informed consent. Both controls and study subjects were normotensive. These subjects were attending medicine OPD for treatment purpose.

Sample size calculation

Where N is sample size for infinite population, Z is z-score at 95% confidence interval (1.96), p is standard deviation (50% or 0.5), M is margin of error (10% or 0.1), n is sample size for finite population and P is size of population (170,000). Approximate sample size calculated using above mentioned formula is 96.

Their fasting blood samples were collected and evaluated for plasma sugar, serum total cholesterol, triglycerides, high density cholesterol, low density cholesterol and very low-density cholesterol. All the serum parameters were analyzed by dry chemistry on Johnson and Johnson Vitros 250. Plasma sugar by GOD-POD method. TC by CHOD-POD method, Tgs by Glcerol kinase -POD method, HDL by CHOD-POD method after precipitation of Non-HDL Lipoproteins with Phosphotungstic acid & MgCl₂, Burstein et al3 and Allain et al4 LDL calculated using Friedewald’s formula. Anthropometric parameters like Height (in Cms), weight (in Kgs) and waist circumference were recorded. BMI was calculated by the formula Body Wt / (Height).2 BFP was calculated by the afore mentioned formula.

Results

Results were expressed as Mean±SD for continuous variables. Statistical analysis included Student’s ‘t’ test done to find statistically significant difference between means, Oneway ANOVA done to assess interrelationship of variables and Pearson’s Correlation Coefficient done to know linear association between variables. The probability value of <.05 at 95% confidence interval considered statistically significant. The biochemical values, BMI and BFP of both cases and controls were as mentioned below. Table 2 (N = Number of subjects)

Mean age was 36.8±9.23 (controls) & 46.92 ±13.66 (DM). Both BMI and BFP showed higher mean±SD in controls compared to cases. Both Controls and Cases fall in Class II obesity group as the mean BMI was 37.07±4.1 for controls and 33.22±8.59 for cases. The mean of Tg between cases & controls was statistically significant (P value <0.001). Increase in VLDL in cases (38.03 ±12.13) compared to controls (29.86±9.99) was observed which was secondary to high Glucose concentrations.

Among the 35 Diabetes Mellitus patients 20 were men and 15 were women, whereas in controls there was female preponderance 30 women as compared to 5 men. The ideal percentage of body fat according to Jackson & Pollard criteria was 19% to 26% for females in the age group of 30 - 55yrs, whereas the same is 13% - 21% in men.

In our study Mean ± SD values of the variables in men & women was as follows.

The Mean±SD of continuous variables and BMI showed marginal variation by way of gender difference, but BFP was moderately elevated in females compared to males (44.35±13.18 in♀ & 33.29±8.97 in ♂) Table 3. This finding was in accordance with the study of Katherine M Flegal, et al 2009.5

BMI and BFP showed statistically significant correlation with lipid profile in both cases and controls with p value <.00001 except HDL in cases, BFP did not show correlation with HDL (P value <.1234).(Table 4) The present study was in agreement with the findings (Imamura et al., 1993) on relationship between fat distribution and serum lipids.

Higher BMI and BFP were detected in controls compared to diabetic cases, though statistically significant correlation was observed (t = 2.389, p=<0.003 BMI) & (t=2.828, p=<0.003 BFP) (Table 2). This study did not correlate with the study of Garg, Deepika Kumar et al. (2019).6 The cases included by Garg et al were basically underweight with BMI of <18.5 kg/m². Similarly, study by Anjana et al. reported that visceral and central abdominal fat was high among diabetic subjects as compared to nondiabetic subjects (P value = 0.005 and 0.011).7 Arora et al. report that the percentage of body fat was significantly high among diabetics as compared to healthy controls (P value < 0.05.8 Strotmeyer et al.9 in their study stated that fat mass was significantly high among individuals with diabetes mellitus (P value < 0.05).

Pearson’s correlation showed negative correlation of BMI & BFP with triglycerides (r = -0.257 & -0.317 respectively). Similarly, LDL showed negative correlation (r = -0,0684) with BFP in diabetic subjects. (Table 6) This finding in our study disproves the results of similar study by Shukohifar, M., Mozafari, Z., et al. (2022).10 Significant negative correlation was observed amongst cases and controls for both BMI & BFP r = - 0.0162 & -0,0336 respectively (Table 6)

The correlation between cases and controls for BMI and BFP was weakly negative (r -0.0162 & -0.0336 respectively) Table 5. The correlation of BMI with BFP in men and women was strongly positive in Diabetic cases (r = 0.914, & 0.977 respectively) Table 6. This finding is in accordance with that of Pihl and Jurimae, 2001.11 In T2DM insulin resistance leads to failure of peripheral utilization of Glucose and increased synthesis of VLDL. Storage of depot fats mainly TGs leads to central obesity which was proved by elevated BF% in our study. The correlation of both BMI & BF% with TC & HDL is weakly positive (r = 0.125 & 0.0303) (r= 0.208, 0.192) and the same with TG is weakly negative (-0.257, -0.319). These findings do not agree with the study of Manu Arora et al (2007).8

Between cases and controls BMI & BF% showed statistically highly significant correlation (P value < 0.00001) Table 7 as assessed by one way ANOVA of differences in means. Though BMI does not discriminate between lean body mass and amount of body fat, in our study we could not gain much value regarding BF% and its correlation with Lipid profile. Prevalence of obesity was almost double as estimated by BF% compared to BMI in a study conducted by Okorodudu D et al. (2010).12 In our study no such conclusion could be achieved. As such we calculated BF% based on BMI, hence accuracy of BF% was not confirmed. Hence, we opined it was wrong to ascertain the superiority of BF% over BMI. In a future study we may calculate BF% by anthropometric methods/ ultrasound / CT scan and compare with calculated BF% to know accuracy of our present findings.

Discussion

As suggested by Jackson & Pollard there was obvious difference in BFP between men and women. In males BFP was less compared to females. The same was proven in our study as shown in Table 3 (44.35±13.18 in♀ & 33.29±8.97 in ♂). BFP is a factor of BMI as shown in the formula, this was constraint in our study. There was strong correlation of BMI and BFP between subjects and controls as shown by one way ANOVA (P < 0.0001) Our study group showed higher BFP in controls (44.9± 7.66) compared to cases (38.03± 12.13) which did not correlate with the study of Vineetha K, Ramdas Nayak et al. 13 These results confirm inclusion of BFP in risk assessment of T2DM in nonobese individuals.14

In Diabetic subjects TC, TG, LDL correlated with BFP but not HDL with p value <0.00001 (Table 4) where as BMI correlated significantly with all lipid parameters. The same was proven in the study conducted in urban population by A Misra, RM Pandey et al. in 2001.15

In our study Tg levels were high in subjects compared to controls whereas HDL-C were low, (Table 1) which was considered to be due to dysregulation of glucose metabolism with insulin resistance and β-cell dysfunction. The hypertriglyceridemia would further aggravate hyperglycaemic state.

Conclusion

Type 2 Diabetes Mellitus is certainly associated with obesity and related complications. It is necessary to design markers of obesity detection to avoid life threatening complications of obesity, these markers of obesity should be cost effective and technically easy to perform. Our study confirms inclusion of BFP along with lipid parameters in T2DM would be fetching in risk assessment.

Limitations of the Study

A major limitation of this study is the small sample size. A study with similar design on a larger scale, involving multiple centres is required to further evaluate the conclusions drawn herewith.

Abbreviations

T2DM = Type 2 Diabetes Mellitus; BMI = Body Mass Index; BFP = Body Fat Percentage; WC= Waist Circumference; WHR = Waist Hip Ratio; ANOVA = Analysis of Variance; TC = Total Cholesterol; TG = Triglycerides; HDL = High Density Cholesterol; LDL = Low Density Cholesterol; VLDL = Very Low-Density Cholesterol

Source of Funding

None.

Conflict of Interest

None.

Acknowledgement

I sincerely thank my colleagues for drafting and analysis of data. My heartfelt thanks to laboratory staff for their assistance. My sincere thanks to Dr. Vijayakumari (MD) for her guidance.